EUVIMEDEUVIMED entdecken
Open Access

Improved Automated Quality Control of Skeletal Wrist Radiographs Using Deep Multitask Learning

Guy Hembroff, Chad Klochko, Joseph Craig, Harikrishnan Changarnkothapeecherikkal, Richard Q Loi

Journal of Imaging Informatics in Medicine · 2024

DOI: 10.1007/s10278-024-01220-9 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging Informatics in Medicine, 2024. Verfasst von Guy Hembroff und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Applied Computing, Michigan Technological University, 1400 Townsend Drive, Houghton, MI, 49931, USA. hembroff@mtu.edu. Hauptthemen laut MEDLINE: Deep Learning, Wrist, Radiography. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 21 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Radiographic quality control is an integral component of the radiology workflow. In this study, we developed a convolutional neural network model tailored for automated quality control, specifically designed to detect and classify key attributes of wrist radiographs including projection, laterality (based on the right/left marker), and the presence of hardware and/or casts. The model's primary objective was to ensure the congruence of results with image requisition metadata to pass the quality assessment. Using a dataset of 6283 wrist radiographs from 2591 patients, our multitask-capable deep learning model based on DenseNet 121 architecture achieved high accuracy in classifying projections (F1 Score of 97.23%), detecting casts (F1 Score of 97.70%), and identifying surgical hardware (F1 Score of 92.27%). The model's performance in laterality marker detection was lower (F1 Score of 82.52%), particularly for partially visible or cut-off markers. This paper presents a comprehensive evaluation of our model's performance, highlighting its strengths, limitations, and the challenges encountered during its development and implementation. Furthermore, we outline planned future research directions aimed at refining and expanding the model's capabilities for improved clinical utility and patient care in radiographic quality control.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningWristRadiographyHumansQuality ControlNeural Networks, Computer

Stichwörter der Autor:innen

Convolutional neural networkDeep learningHealthcare quality improvementImage analysisImage classificationRadiographic quality control

Autor:innen und Institutionen

  • Department of Applied Computing, Michigan Technological University, 1400 Townsend Drive, Houghton, MI, 49931, USA. hembroff@mtu.edu.
  • Department of Radiology, Division of Musculoskeletal Radiology, Henry Ford Hospital, 2799 West Grand Boulevard, Detroit, MI, 48202, USA.
  • Department of Applied Computing, Michigan Technological University, 1400 Townsend Drive, Houghton, MI, 49931, USA.

Zitieren

APA

Hembroff, G., Klochko, C., Craig, J., Changarnkothapeecherikkal, H., & Loi, R. Q. (2024). Improved Automated Quality Control of Skeletal Wrist Radiographs Using Deep Multitask Learning. Journal of Imaging Informatics in Medicine. https://doi.org/10.1007/s10278-024-01220-9

Vancouver

Hembroff G, Klochko C, Craig J, Changarnkothapeecherikkal H, Loi RQ. Improved Automated Quality Control of Skeletal Wrist Radiographs Using Deep Multitask Learning. Journal of Imaging Informatics in Medicine. 2024. doi:10.1007/s10278-024-01220-9

BibTeX
@article{Hembroff2024,
  author = {Hembroff, Guy and Klochko, Chad and Craig, Joseph and Changarnkothapeecherikkal, Harikrishnan and Loi, Richard Q},
  title = {Improved Automated Quality Control of Skeletal Wrist Radiographs Using Deep Multitask Learning},
  journal = {Journal of Imaging Informatics in Medicine},
  year = {2024},
  doi = {10.1007/s10278-024-01220-9}
}
Kostenlos starten →